Bee functional traits and their relationship to pollination services depend on many factors: A meta-regression analysis

Understanding relationships between organisms and the ecosystem services they provide is crucial for predicting the impacts of continued biodiversity changes. Functional traits of organisms can affect ecosystem properties and are thus increasingly used to predict long-term ecosystem functioning. Bees are ideal taxa for using functional approaches given their role in pollination for many plant species and wide diversity of traits. Although distributions of bee functional traits are being documented in the literature, there is a clear lack of understanding of how they relate to ecosystem functioning (i.e. pollination). To address this knowledge gap, we conducted a meta-regression analysis with the following objectives: (1) quantify the effects of bee functional traits on pollination and (2) assess sources of heterogeneity to identify variables that might explain variation across studies. Seventeen studies met our criteria for inclusion, yielding 45 individual effect sizes for six traits (body size, tongue length, diet breadth, nesting, parasitism, and sociality). Overall, bee functional traits had a significant effect on pollination; however, effect sizes were weak with high variability across studies. Sources of heterogeneity included the metrics used to quantify pollination, the number of bee genera analysed in a study, and whether traits were measured categorically or continuously. These results indicate a need for more research to improve understanding of trait–pollination relationships. For some bee traits, effects on pollination may be dependent on other factors like bee abundance, environmental context, and the plant species evaluated in studies.